Total Peripheral Parenteral Nutrition in Pregnancy LYDIA A.
WATSON, M.D., ALEXANDER A. BOMMARITO, R.PH.,
AND
JOSEPH F. MARSHALL, M.D.
From the Department of Obstetrics and Gynecology, Saginaw Cooperative Hospitals, Inc.; Department of Obstetrics, Biology and Department of Surgery, Michigan State University College of Human Medicine, and Saginaw General
ABSTRACT. Twenty pregnant patients needing nutritional support for various indications received hypercaloric, hyperos-
motic, "3
in 1," peripheral parenteral nutrition as a bridge to enteral therapy. This system, named total peripheral parenteral nutrition, was evaluated as to tolerance and efficacy in pregnant patients. Patients were in various stages of pregnancy and had an average weight loss of 10.4 pounds prior to admission. Patients were maintained on total peripheral parenteral nutrition for an average of 5 days and gained an average of 4.1 pounds. Good tolerance with minimal side effects of the treat-
Gynecology, and Reproductive Hospital, Saginaw, Michigan
noted. Intravenous sites were changed an average of patient during the course of therapy and only one serious complication was noted in 20 patients. Patients were followed through delivery and this information is presented. Total peripheral parenteral nutrition appears to be an acceptable alternative to conventional total parenteral nutrition to pregnant patients needing nutritional support. These hypertonic solutions can supply total caloric/metabolic needs without Journal of Parenteral and Enteral unacceptable side effects. ( Nutrition 14: 485-489, 1990)
ment was
1.1 times per
Since 1968 when Dudrick et all first adopted the and home hyperalimentation in eight hyperemesis gravtechnique of subclavian vein catheterization for hyper- idarum patients with severe nutritional depletion led us alimentation, total parenteral nutrition has become an to believe that identifying deficits earlier and treating acceptable therapy in all fields for patients who cannot such patients with a system associated with fewer risks tolerate enteral nutrition. In contrast to other medical/ could lead to safer, more cost-effective care. Reformulasurgical specialists, obstetricians have used total paren- tion of standard solutions to a higher pH, along with the teral nutrition less frequently. Unknown fetal conse- addition of fat to lower the colligative properties and increase the buffering effect on the vein, brought about quences, possible maternal metabolic problems, risk of sepsis, and inherent complications with central venous the idea that the &dquo;3 in 1&dquo; solutions may be administered catheter placement contribute to total parenteral nutri- peripherally with higher concentrations of carbohydrates tion’s use as a &dquo;last resort&dquo; therapy. However, in obstet- to supply more calories. Successful experimentation with rics, parenteral nutrition may be vital to restore and &dquo;3 in 1&dquo; on a peripheral basis with hyperosmolar solumaintain maternal nutritional status whenever pro- tions for over 1000 days of therapy in surgical and longed nutritional deprivation exists. Because nutrition- medical patients at our institution motivated us to unally deficient gravid women encounter significantly in- dertake this study in pregnant patients. To our knowlcreased risks of dehydration, vitamin deficiencies, elec- edge, peripheral parenteral nutrition with hyperosmolar trolyte imbalances, delivery of low birth weight infants.~°3 solutions has never been successfully accomplished. The purpose of this study is to investigate the tolerance and even fetal demise,4°5 hyperalimentation should be and efficacy of hypercaloric, hyperosmotic, &dquo;all in one,&dquo; considered more often. Advances in the field of nutrition may offer new peripheral parenteral nutrition as a bridge to enteral choices for obstetricians. New nutrient mixtures have therapy in pregnant patients needing nutritional supbeen devised, including the &dquo;3 in 1&dquo; system which com- port. Henceforth this system shall be referred to as total bines amino acids, glucose, and fats in one bag.6 Advan- peripheral parenteral nutrition. tages include ease of administration and preparation, MATERIALS AND METHODS decreased potential for contamination during manipulation, and cost savings. Another proposed approach has This study was approved by the institutional review been to administer nutritional support through periph- board at Saginaw General Hospital and informed consent eral veins (peripheral parenteral nutrition).&dquo; This mode was obtained before entrance into the study. Between of hyperalimentation avoids the risks of central vein January 1988 and December 1988, 20 patients were incannulation, facilitates infection control, reduces nursing cluded in the study. Seventeen patients were in the first
time, and allows for earlier nutritional intervention.
However, solutions used previously on a peripheral basis were rarely able to supply total caloric/metabolic needs without unacceptable side effects.8.9 Our prior experience with total parenteral nutrition Reprint requests: Mr. Alexander A. Bommarito, Saginaw Hospital, 1447 N. Harrison, Saginaw. MI 48602.
General
trimester of pregnancy, one in the second trimester, and two in the third trimester. Inclusion criteria consisted of inability to use the gastrointestinal tract, nothing by mouth status for greater than 72 hr. or inadequate intake with moderate to severe nutritional depletion. Excluded were patients in need of long-term (greater than 21 days) nutritional support. Patient population included 16 patients with hyperemesis gravidarum and dehydration (with one or more of
485 TECH UNIV LIBRARY on May 22, 2015 Downloaded from pen.sagepub.com at NANYANG
486
supplied as &dquo;multi-Trace.5&dquo; (1 ml contammg Zn 1 mg, Cu 0.4 mg, Mn 0.1 mg, Cr 4 Ag, Se 20 gg). Iron was added (1-3 mg) based on serum iron values. Heparin, 1000-2000 U/1500 ml, was included to decrease vein irritation. Additionally, patients received 10 ml of mul-
the following: increased urine specitic gravity, ketonuria, orthostatic changes); a patient with active Chron’s disease and a small bowel obstruction; a patient with sep-
were
recurrent pyelonephritis, and moderate nutridepletion; a patient with a pancreatic pseudocyst requiring pancreatic cyst jejunostomy; and a patient admitted with nonspecific abdominal pain of questionable gastrointestinal tract etiology and poor nutritional intake. Three patients had been hospitalized once previously during their pregnancy and two patients had three prior hospitalizations. Thirteen patients had received outpatient treatment [oral or rectal antiemetics; one patient received intravenous (IV) hydration in her physician’s office] without relief of symptoms. Concurrent urinary tract infections were detected in two patients and one patient had an accompanying acute bronchitis. The average patient age was 25.5 years old. Average gravidity and parity were 2.8 and 1.1, respectively. Nine of 13 nulliparous women had histories of hyperemesis with previous pregnancies. The average weight loss prior to admission was 10.4 pounds. The average weight deficit based on expected maternal weight gain for gestational age was 13.7 pounds.
ticemia, tional
tivitamins (MVI-12)
daily.
Pharmacy Preparation Pharmacists were responsible for bag preparation using standard aseptic technique. The order of mixing was strictly controlled along manufacturer’s guidelines.&dquo; Compatibilities with other medications were checked. Proper labeling of the bags was performed prior to delivery to the floor. The
stability
of solutions and additives
monitored, and &dquo;wasted&dquo; total peripheral parenteral nutrition bags (eg, physician order changes) were docu-
was
mented and held to a minimum. Admixtures used in this study were Abbott’s Aminosyn II, Liposyn II, and dextrose
injection.
Nursing Care All mixtures were allowed to hang for up to 48 hr and infused through a single volumetric pump. Piggyback medications were allowed when compatible. Nurses inserted iv cannulae (intracatheters when possible ranging from 20-16 gauge) using sterile techniques after cleansing skin with Betadine and alcohol. Sterile dressings were applied to the iv site and changed every 48 hr. Sites were monitored hourly and discontinued for pain, redness, or swelling. The number of iv sites during treatment and the indications for removal were documented were
Total
Peripheral Parenteral Nutrition Design
All patients received an &dquo;all in one&dquo; formulation through a large peripheral arm vein (wrist or forearm) after determination of resting energy equivalent and respiratory quotient values by indirect calorimetry. Patients received 1424-2600 kcal/24 hr, based on resting energy equivalent values plus 20% plus 300 calories, in accordance with recommended pregnancy requirements. Protein needs were calculated at 1.0-1.5 g/kg. Mixtures of 20-30% dextrose, 7 or 10% Aminosyn II, and 10 or 20% Liposyn II provided the necessary calories and protein. Solutions were started at 30 ml/hr and increased to 60 ml/hr in another 4 hr. Further adjustments of rate
based
(Table II). Standard Monitoring
energy needs. solutions used 20% dextrose in the &dquo;3 in 1&dquo; formulation to lower initial osmolarity. It is important to note that large amounts of potassium were required to replace the catabolic losses and additional pregnancy requirements. In addition, potassium is a well-known vein irritant and can be the limiting factor in acceptable total peripheral parenteral nutrition. Therefore, the lower osmolarity and potassium needs of a 20% dextrose solution provided for a safer and more effective initiation of total peripheral parenteral nutrition. Patients were’ then advanced to a 30% dextrose mixture after potassiumL deficits were replaced. This program provided a more~ favorable metabolic base for both mother and fetus. In our experience with total parenteral nutrition, highfat feedings keep patients ketotic and high concentrations of carbohydrates decrease symptoms of nausea ancI vomiting. The base solution of amino acids 10%, dextrosfi 30%, and fat 10% (1000 cc of each) provides the necessaryr nonketotic response. The osmolarity of this base solutiori is 836 mOsm/liter with 0.84 cal/ml. Approximate calorie;1 per feeding rate are listed in Table I. Electrolytes and minerals were added based on pa tients’ needs using standard order forms. Trace element:s
were
on
Start-up
’
’
.
-
Monitoring included daily intake, urine output, weight (performed on the same Healthometer scale in hospital gown only), SMA-23 days 1 and 4 of therapy then weekly, electrolytes daily for 3 days then as needed, dry diet as tolerated, calorie counts, serum iron and iron binding capacity levels, fingerstick blood sugar levels every 6 hr for the first 24 hr, and metabolic studies preintervention and 3 to 5 days postintervention. All patients with the exception of the patient with the bowel obstruction were started on dry diet on admission. Total peripheral parenteral nutrition was continued until adequate oral nutrition was established and verified by calorie counts. Total peripheral parenteral nutrition was either discontinued as soon as appropriate calorie counts were noted or weaned as gradual increases in oral calories were noted. RESULTS
Initial resting energy equivalents ranged from 8642491 kcal/24 hr with respiratory quotients of 0.64-0.82. All patients were hypokalemic or in the low normal range (2.8-3.9) on admission while serum albumin levels ranged from 2.6-4.3 and triglyceride levels ranged from 36-263. Iron deficiency was noted in eight patients. The iv sites were changed an average of 1.1 times per
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
487 TABLE I Standard formulation
TPPN,
total
peripheral parenteral nutrition.
TABLE II Intravenous site evaluation
of poor access to peripheral veins and questionable duration of therapy on her third readmission. Seventeen patients (three lost to follow-up because of moving to different states) were followed through pregnancy and delivery. All delivered appropriate-for-gestational-age infants at gestational ages between 361/2 and 411/2 weeks (Table III). The average gestational age was 39 weeks. Eleven patients underwent spontaneous vaginal deliveries, three had repeat cesarean sections, one patient underwent primary cesarean section for cephalopelvic disproportion, one patient had a midforceps delivery, and one had a low forceps delivery for obstetrical indications. Birth weights ranged from 2400-4170 g with an average of 3388 g. One infant had an umbilical hernia and two infants were treated for neonatal physiologic jaundice. None of these abnormalities were attributed to total peripheral parenteral nutrition. Two patients had deliveries at 361/2 weeks’ gestation. No patients underwent tocolysis for preterm labor. No complications unique to pregnancy or total peripheral parenteral nutrition were encountered.
cause
DISCUSSION
Enthusiasm for the nancy
might
use of hyperalimentation in pregincrease if it could be shown to be safe,
TABLE III
Delivery information *
D/C, removal.
patient, with the
most
frequent indication being patient
discomfort. One patient developed a temperature of 101.2° F on the 15th day of total peripheral parenteral nutrition attributable to the iv site, which had remained in site for 6 days. The patient did not appear septic and was treated with an oral cephalosporin antibiotic after removal of the catheter. The patient’s fever resolved within 24 hr. Subsequently, two of three blood cultures grew out Staphylococcus epidermidis, with the catheter tip demonstrating the same organism. Maternal weight gain while on total peripheral parenteral nutrition averaged 4.1 pounds (range 0-8.5) per patient. Total peripheral parenteral nutrition was administered for an average of 5 days (range 1-17 days). No complications unique to pregnancy were encountered. Five patients were rehospitalized (1-3 times) after initial improvement and discharge. They were all restarted on total peripheral parenteral nutrition. One patient did require central venous catheterization be-
I
.
Information
not available out of state.
quently delivered
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
for three
patients because they subse-
488
simple, therapeutic, cost-effective, and readily available all physicians (including those not skilled in central line placement). Our study demonstrates the safety, sim-
to
plicity, and cost-effectiveness of total peripheral parenteral nutrition. Further studies to determine therapeutic efficacy are needed. In our 20 patients, weight deficits prior to admission (Table IV) were considered significant especially for those in the first trimester. Acute protein-energy deficits, dehydration, and hypokalemia were demonstrated through history, physical examination, and laboratory studies. Indirect calorimetry indicated that our patients’ caloric utilization consisted mainly of fats, with respiratory quotients consistently showing a catabolic state. Patients were placed on total peripheral parenteral nutrition to replenish depleted electrolyte and nutrient stores and to provide positive protein and energy balances while awaiting the return of adequate gastrointestinal function. With this type of nutritional support, patients demonstrated improvements in serum levels of albumin, triglycerides, and potassium as well as weight gain. No hyperlipidemia was encountered. We feel that it is especially important to identify patients early in the course of nutritional depletion before their deficits and potassium losses are so great that peripheral nutrition would no longer be an option. Calculation of potassium replacement in the first 48 hr of total peripheral parenteral nutrition in patients revealed an average need of 88 mEq potassium the first 24 hr and 65 mEq potassium the second 24 hr. A recent study suggests that protein-deficient diets may lead to renal function changes and inevitably to decreased renal clearance of various compounds/ drugs. 11 This is yet another reason for identifying protein deficiencies early and restoring adequate nutrition with total peripheral parenteral nutrition. Our system is simple-easier and quicker to prepare and administer, requiring less nursing care and thereby
more cost-effective than traditional total parenteral nutrition. In our institution the patient cost bag runs $90-130, with each bag lasting from 15-48 hr. Actual pharmacy cost comparison of total parenteral nutrition with total peripheral parenteral nutrition per day is listed in Table V. Serious complications associated with central venous catheterization such as pneumothorax, air embolism, and cardiac perforation were avoided. Sepsis was minimized. The incidence of phlebitis with total peripheral parenteral nutrition was 5% and posed no significant morbidity. It is also important to note that the one complication that we encountered might have been avoided if catheter sites had been changed automatically after 3 or 4 days. The major side effect associated with total peripheral parenteral nutrition is phlebitis. Insertion technique, size of the vein, pH of the solution, osmolarity, and medications infused all play a role in the likelihood of developing phlebitis. It has been shown that with an osmolarity greater than 600 mOsm, the incidence of peripheral vein phlebitis is 70%.12 Our solutions have osmolarities greater than 600 (range with additives 1200-1350 mOsm/ liter) but are thought to be less irritating for several
being
reasons.
An increase in the pH to 5.7 decreases irritation and allows fats to be stable in solution with proteins. Fats also have both a dilutional effect because they bear no colligative properties and a buffering effect by coating the vein wa11.13 Heparin has also demonstrated a protective effect in several studies. It is thought to prevent phlebitis by affecting ion charges in the vein wall, thereby
setting up a protective hydrostatic barrier.&dquo; Heparin may also prevent establishment of fibrin, which acts as a nidus for bacterial accumulation.15 The combination of increased pH, fats, and heparin are most likely additive in preventing phlebitis. Subjectively our patients &dquo;feel better&dquo; on total peripheral parenteral nutrition. There also was a decrease in
TABLE IV
Weight statistics (in pounds)
*
Weight during pregnancy before
deficient if different than
predicted.
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
489 TABLE V Cost comparison
balance. Surgery 64:134-142, 1968 2. Smith CA: Effects of maternal undernutrition upon newborn infants in Holland: 1944-1945. J Pediatr 30:229-243, 1947 3. Antonov- AN: Children born during the siege of Leningrad in 1942. J Pediatr 30:250-259. 1947 4. Starks GC: Pregnancy-induced hyperemesis: A reassessment of therapy and proposal of a new etiologic theory. Miss Med 81 :253-
256, 1984
TPN, total parenteral nutrition.
the amount of antiemetics used by these patients as compared with patients treated solely with iv fluids and antiemetics. These may be two areas of further investigation in the future. Patients were followed through delivery; there were no adverse maternal or fetal effects. In conclusion, total peripheral parenteral nutrition offers major advantages over conventional total parenteral nutrition to pregnant patients needing nutritional support. It is well tolerated, has minimal side effects, and is simple and cost-effective. Our study demonstrates that providing hypercaloric, hyperosmolar solutions via peripheral veins over short periods of time is feasible, especially in pregnant patients. Further studies using a control group need to be done to evaluate clinical effectiveness with respect to complications and subsequent pregnancy outcome. REFERENCES 1. Dudrick
SJ, Wilmore DN, Vars HM, et al: Long term total parenteral nutrition with growth development, and positive nitrogen
5. Stellato TA, Danziger LH, Burkons D. et al: Fetal salvage with maternal total parenteral nutrition: The pregnant mother as her own control. JPEN 12:412-413, 1988 6. O’Keefe SJD, Bean E, Symmonds K. et al: Clinical evaluation of a "3 in 1" intravenous nutrient solution. S Afr J Med 68:82-86, 1985 7. Wilson A, Goode AW, Kirk CJC, et al: Parenteral nutrition via peripheral veins: A feasibility study. J R Soc Med 80:430-433, 1987 8. Bodky A: Parenteral nutrition by peripheral vein central venous
9.
catheter or portal vein? World J Surg 10 :47-52, 1986 Gassanega A, Day AT, Sankary H: Efficacy of a 20% fat emulsion administered as a peripherally administered substrate. Surg Gynecol Obstet 160:387-392, 1985.
10.
11. 12.
Sayeed FA, Tripp MG, Sukumaran KB, et al: Stability of various total nutrient admixture formulations using Liposyn II and Aminosyn II. Am J Hosp Pharm 44:2280-2286, 1987 Flanagan S: Research highlights. Research Resources Reporter Feb:11-13, 1988 Gazitria R, Wilson K, Bistrain BR, et al: Factors determining peripheral vein tolerance to amino acid infusions. Arch Surg 114:897-900, 1979
13.
Fujiwara T, Kawarasaki H, Fonkalsrud E: Reduction of postinfusion venous endothelial injury with intralipid. Surg Gynecol
Obstet 158:57-65, 1984 14. Tanner WA, Delaney PV, Hennessey TP: The influence of heparin on intravenous infusions. Br J Surg 67:311-312, 1980 15. Maki DG: Preventing infection in intravenous therapy. Anesth Analg 56:141-153, 1977
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
WATSON, M.D., ALEXANDER A. BOMMARITO, R.PH.,
AND
JOSEPH F. MARSHALL, M.D.
From the Department of Obstetrics and Gynecology, Saginaw Cooperative Hospitals, Inc.; Department of Obstetrics, Biology and Department of Surgery, Michigan State University College of Human Medicine, and Saginaw General
ABSTRACT. Twenty pregnant patients needing nutritional support for various indications received hypercaloric, hyperos-
motic, "3
in 1," peripheral parenteral nutrition as a bridge to enteral therapy. This system, named total peripheral parenteral nutrition, was evaluated as to tolerance and efficacy in pregnant patients. Patients were in various stages of pregnancy and had an average weight loss of 10.4 pounds prior to admission. Patients were maintained on total peripheral parenteral nutrition for an average of 5 days and gained an average of 4.1 pounds. Good tolerance with minimal side effects of the treat-
Gynecology, and Reproductive Hospital, Saginaw, Michigan
noted. Intravenous sites were changed an average of patient during the course of therapy and only one serious complication was noted in 20 patients. Patients were followed through delivery and this information is presented. Total peripheral parenteral nutrition appears to be an acceptable alternative to conventional total parenteral nutrition to pregnant patients needing nutritional support. These hypertonic solutions can supply total caloric/metabolic needs without Journal of Parenteral and Enteral unacceptable side effects. ( Nutrition 14: 485-489, 1990)
ment was
1.1 times per
Since 1968 when Dudrick et all first adopted the and home hyperalimentation in eight hyperemesis gravtechnique of subclavian vein catheterization for hyper- idarum patients with severe nutritional depletion led us alimentation, total parenteral nutrition has become an to believe that identifying deficits earlier and treating acceptable therapy in all fields for patients who cannot such patients with a system associated with fewer risks tolerate enteral nutrition. In contrast to other medical/ could lead to safer, more cost-effective care. Reformulasurgical specialists, obstetricians have used total paren- tion of standard solutions to a higher pH, along with the teral nutrition less frequently. Unknown fetal conse- addition of fat to lower the colligative properties and increase the buffering effect on the vein, brought about quences, possible maternal metabolic problems, risk of sepsis, and inherent complications with central venous the idea that the &dquo;3 in 1&dquo; solutions may be administered catheter placement contribute to total parenteral nutri- peripherally with higher concentrations of carbohydrates tion’s use as a &dquo;last resort&dquo; therapy. However, in obstet- to supply more calories. Successful experimentation with rics, parenteral nutrition may be vital to restore and &dquo;3 in 1&dquo; on a peripheral basis with hyperosmolar solumaintain maternal nutritional status whenever pro- tions for over 1000 days of therapy in surgical and longed nutritional deprivation exists. Because nutrition- medical patients at our institution motivated us to unally deficient gravid women encounter significantly in- dertake this study in pregnant patients. To our knowlcreased risks of dehydration, vitamin deficiencies, elec- edge, peripheral parenteral nutrition with hyperosmolar trolyte imbalances, delivery of low birth weight infants.~°3 solutions has never been successfully accomplished. The purpose of this study is to investigate the tolerance and even fetal demise,4°5 hyperalimentation should be and efficacy of hypercaloric, hyperosmotic, &dquo;all in one,&dquo; considered more often. Advances in the field of nutrition may offer new peripheral parenteral nutrition as a bridge to enteral choices for obstetricians. New nutrient mixtures have therapy in pregnant patients needing nutritional supbeen devised, including the &dquo;3 in 1&dquo; system which com- port. Henceforth this system shall be referred to as total bines amino acids, glucose, and fats in one bag.6 Advan- peripheral parenteral nutrition. tages include ease of administration and preparation, MATERIALS AND METHODS decreased potential for contamination during manipulation, and cost savings. Another proposed approach has This study was approved by the institutional review been to administer nutritional support through periph- board at Saginaw General Hospital and informed consent eral veins (peripheral parenteral nutrition).&dquo; This mode was obtained before entrance into the study. Between of hyperalimentation avoids the risks of central vein January 1988 and December 1988, 20 patients were incannulation, facilitates infection control, reduces nursing cluded in the study. Seventeen patients were in the first
time, and allows for earlier nutritional intervention.
However, solutions used previously on a peripheral basis were rarely able to supply total caloric/metabolic needs without unacceptable side effects.8.9 Our prior experience with total parenteral nutrition Reprint requests: Mr. Alexander A. Bommarito, Saginaw Hospital, 1447 N. Harrison, Saginaw. MI 48602.
General
trimester of pregnancy, one in the second trimester, and two in the third trimester. Inclusion criteria consisted of inability to use the gastrointestinal tract, nothing by mouth status for greater than 72 hr. or inadequate intake with moderate to severe nutritional depletion. Excluded were patients in need of long-term (greater than 21 days) nutritional support. Patient population included 16 patients with hyperemesis gravidarum and dehydration (with one or more of
485 TECH UNIV LIBRARY on May 22, 2015 Downloaded from pen.sagepub.com at NANYANG
486
supplied as &dquo;multi-Trace.5&dquo; (1 ml contammg Zn 1 mg, Cu 0.4 mg, Mn 0.1 mg, Cr 4 Ag, Se 20 gg). Iron was added (1-3 mg) based on serum iron values. Heparin, 1000-2000 U/1500 ml, was included to decrease vein irritation. Additionally, patients received 10 ml of mul-
the following: increased urine specitic gravity, ketonuria, orthostatic changes); a patient with active Chron’s disease and a small bowel obstruction; a patient with sep-
were
recurrent pyelonephritis, and moderate nutridepletion; a patient with a pancreatic pseudocyst requiring pancreatic cyst jejunostomy; and a patient admitted with nonspecific abdominal pain of questionable gastrointestinal tract etiology and poor nutritional intake. Three patients had been hospitalized once previously during their pregnancy and two patients had three prior hospitalizations. Thirteen patients had received outpatient treatment [oral or rectal antiemetics; one patient received intravenous (IV) hydration in her physician’s office] without relief of symptoms. Concurrent urinary tract infections were detected in two patients and one patient had an accompanying acute bronchitis. The average patient age was 25.5 years old. Average gravidity and parity were 2.8 and 1.1, respectively. Nine of 13 nulliparous women had histories of hyperemesis with previous pregnancies. The average weight loss prior to admission was 10.4 pounds. The average weight deficit based on expected maternal weight gain for gestational age was 13.7 pounds.
ticemia, tional
tivitamins (MVI-12)
daily.
Pharmacy Preparation Pharmacists were responsible for bag preparation using standard aseptic technique. The order of mixing was strictly controlled along manufacturer’s guidelines.&dquo; Compatibilities with other medications were checked. Proper labeling of the bags was performed prior to delivery to the floor. The
stability
of solutions and additives
monitored, and &dquo;wasted&dquo; total peripheral parenteral nutrition bags (eg, physician order changes) were docu-
was
mented and held to a minimum. Admixtures used in this study were Abbott’s Aminosyn II, Liposyn II, and dextrose
injection.
Nursing Care All mixtures were allowed to hang for up to 48 hr and infused through a single volumetric pump. Piggyback medications were allowed when compatible. Nurses inserted iv cannulae (intracatheters when possible ranging from 20-16 gauge) using sterile techniques after cleansing skin with Betadine and alcohol. Sterile dressings were applied to the iv site and changed every 48 hr. Sites were monitored hourly and discontinued for pain, redness, or swelling. The number of iv sites during treatment and the indications for removal were documented were
Total
Peripheral Parenteral Nutrition Design
All patients received an &dquo;all in one&dquo; formulation through a large peripheral arm vein (wrist or forearm) after determination of resting energy equivalent and respiratory quotient values by indirect calorimetry. Patients received 1424-2600 kcal/24 hr, based on resting energy equivalent values plus 20% plus 300 calories, in accordance with recommended pregnancy requirements. Protein needs were calculated at 1.0-1.5 g/kg. Mixtures of 20-30% dextrose, 7 or 10% Aminosyn II, and 10 or 20% Liposyn II provided the necessary calories and protein. Solutions were started at 30 ml/hr and increased to 60 ml/hr in another 4 hr. Further adjustments of rate
based
(Table II). Standard Monitoring
energy needs. solutions used 20% dextrose in the &dquo;3 in 1&dquo; formulation to lower initial osmolarity. It is important to note that large amounts of potassium were required to replace the catabolic losses and additional pregnancy requirements. In addition, potassium is a well-known vein irritant and can be the limiting factor in acceptable total peripheral parenteral nutrition. Therefore, the lower osmolarity and potassium needs of a 20% dextrose solution provided for a safer and more effective initiation of total peripheral parenteral nutrition. Patients were’ then advanced to a 30% dextrose mixture after potassiumL deficits were replaced. This program provided a more~ favorable metabolic base for both mother and fetus. In our experience with total parenteral nutrition, highfat feedings keep patients ketotic and high concentrations of carbohydrates decrease symptoms of nausea ancI vomiting. The base solution of amino acids 10%, dextrosfi 30%, and fat 10% (1000 cc of each) provides the necessaryr nonketotic response. The osmolarity of this base solutiori is 836 mOsm/liter with 0.84 cal/ml. Approximate calorie;1 per feeding rate are listed in Table I. Electrolytes and minerals were added based on pa tients’ needs using standard order forms. Trace element:s
were
on
Start-up
’
’
.
-
Monitoring included daily intake, urine output, weight (performed on the same Healthometer scale in hospital gown only), SMA-23 days 1 and 4 of therapy then weekly, electrolytes daily for 3 days then as needed, dry diet as tolerated, calorie counts, serum iron and iron binding capacity levels, fingerstick blood sugar levels every 6 hr for the first 24 hr, and metabolic studies preintervention and 3 to 5 days postintervention. All patients with the exception of the patient with the bowel obstruction were started on dry diet on admission. Total peripheral parenteral nutrition was continued until adequate oral nutrition was established and verified by calorie counts. Total peripheral parenteral nutrition was either discontinued as soon as appropriate calorie counts were noted or weaned as gradual increases in oral calories were noted. RESULTS
Initial resting energy equivalents ranged from 8642491 kcal/24 hr with respiratory quotients of 0.64-0.82. All patients were hypokalemic or in the low normal range (2.8-3.9) on admission while serum albumin levels ranged from 2.6-4.3 and triglyceride levels ranged from 36-263. Iron deficiency was noted in eight patients. The iv sites were changed an average of 1.1 times per
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
487 TABLE I Standard formulation
TPPN,
total
peripheral parenteral nutrition.
TABLE II Intravenous site evaluation
of poor access to peripheral veins and questionable duration of therapy on her third readmission. Seventeen patients (three lost to follow-up because of moving to different states) were followed through pregnancy and delivery. All delivered appropriate-for-gestational-age infants at gestational ages between 361/2 and 411/2 weeks (Table III). The average gestational age was 39 weeks. Eleven patients underwent spontaneous vaginal deliveries, three had repeat cesarean sections, one patient underwent primary cesarean section for cephalopelvic disproportion, one patient had a midforceps delivery, and one had a low forceps delivery for obstetrical indications. Birth weights ranged from 2400-4170 g with an average of 3388 g. One infant had an umbilical hernia and two infants were treated for neonatal physiologic jaundice. None of these abnormalities were attributed to total peripheral parenteral nutrition. Two patients had deliveries at 361/2 weeks’ gestation. No patients underwent tocolysis for preterm labor. No complications unique to pregnancy or total peripheral parenteral nutrition were encountered.
cause
DISCUSSION
Enthusiasm for the nancy
might
use of hyperalimentation in pregincrease if it could be shown to be safe,
TABLE III
Delivery information *
D/C, removal.
patient, with the
most
frequent indication being patient
discomfort. One patient developed a temperature of 101.2° F on the 15th day of total peripheral parenteral nutrition attributable to the iv site, which had remained in site for 6 days. The patient did not appear septic and was treated with an oral cephalosporin antibiotic after removal of the catheter. The patient’s fever resolved within 24 hr. Subsequently, two of three blood cultures grew out Staphylococcus epidermidis, with the catheter tip demonstrating the same organism. Maternal weight gain while on total peripheral parenteral nutrition averaged 4.1 pounds (range 0-8.5) per patient. Total peripheral parenteral nutrition was administered for an average of 5 days (range 1-17 days). No complications unique to pregnancy were encountered. Five patients were rehospitalized (1-3 times) after initial improvement and discharge. They were all restarted on total peripheral parenteral nutrition. One patient did require central venous catheterization be-
I
.
Information
not available out of state.
quently delivered
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
for three
patients because they subse-
488
simple, therapeutic, cost-effective, and readily available all physicians (including those not skilled in central line placement). Our study demonstrates the safety, sim-
to
plicity, and cost-effectiveness of total peripheral parenteral nutrition. Further studies to determine therapeutic efficacy are needed. In our 20 patients, weight deficits prior to admission (Table IV) were considered significant especially for those in the first trimester. Acute protein-energy deficits, dehydration, and hypokalemia were demonstrated through history, physical examination, and laboratory studies. Indirect calorimetry indicated that our patients’ caloric utilization consisted mainly of fats, with respiratory quotients consistently showing a catabolic state. Patients were placed on total peripheral parenteral nutrition to replenish depleted electrolyte and nutrient stores and to provide positive protein and energy balances while awaiting the return of adequate gastrointestinal function. With this type of nutritional support, patients demonstrated improvements in serum levels of albumin, triglycerides, and potassium as well as weight gain. No hyperlipidemia was encountered. We feel that it is especially important to identify patients early in the course of nutritional depletion before their deficits and potassium losses are so great that peripheral nutrition would no longer be an option. Calculation of potassium replacement in the first 48 hr of total peripheral parenteral nutrition in patients revealed an average need of 88 mEq potassium the first 24 hr and 65 mEq potassium the second 24 hr. A recent study suggests that protein-deficient diets may lead to renal function changes and inevitably to decreased renal clearance of various compounds/ drugs. 11 This is yet another reason for identifying protein deficiencies early and restoring adequate nutrition with total peripheral parenteral nutrition. Our system is simple-easier and quicker to prepare and administer, requiring less nursing care and thereby
more cost-effective than traditional total parenteral nutrition. In our institution the patient cost bag runs $90-130, with each bag lasting from 15-48 hr. Actual pharmacy cost comparison of total parenteral nutrition with total peripheral parenteral nutrition per day is listed in Table V. Serious complications associated with central venous catheterization such as pneumothorax, air embolism, and cardiac perforation were avoided. Sepsis was minimized. The incidence of phlebitis with total peripheral parenteral nutrition was 5% and posed no significant morbidity. It is also important to note that the one complication that we encountered might have been avoided if catheter sites had been changed automatically after 3 or 4 days. The major side effect associated with total peripheral parenteral nutrition is phlebitis. Insertion technique, size of the vein, pH of the solution, osmolarity, and medications infused all play a role in the likelihood of developing phlebitis. It has been shown that with an osmolarity greater than 600 mOsm, the incidence of peripheral vein phlebitis is 70%.12 Our solutions have osmolarities greater than 600 (range with additives 1200-1350 mOsm/ liter) but are thought to be less irritating for several
being
reasons.
An increase in the pH to 5.7 decreases irritation and allows fats to be stable in solution with proteins. Fats also have both a dilutional effect because they bear no colligative properties and a buffering effect by coating the vein wa11.13 Heparin has also demonstrated a protective effect in several studies. It is thought to prevent phlebitis by affecting ion charges in the vein wall, thereby
setting up a protective hydrostatic barrier.&dquo; Heparin may also prevent establishment of fibrin, which acts as a nidus for bacterial accumulation.15 The combination of increased pH, fats, and heparin are most likely additive in preventing phlebitis. Subjectively our patients &dquo;feel better&dquo; on total peripheral parenteral nutrition. There also was a decrease in
TABLE IV
Weight statistics (in pounds)
*
Weight during pregnancy before
deficient if different than
predicted.
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
489 TABLE V Cost comparison
balance. Surgery 64:134-142, 1968 2. Smith CA: Effects of maternal undernutrition upon newborn infants in Holland: 1944-1945. J Pediatr 30:229-243, 1947 3. Antonov- AN: Children born during the siege of Leningrad in 1942. J Pediatr 30:250-259. 1947 4. Starks GC: Pregnancy-induced hyperemesis: A reassessment of therapy and proposal of a new etiologic theory. Miss Med 81 :253-
256, 1984
TPN, total parenteral nutrition.
the amount of antiemetics used by these patients as compared with patients treated solely with iv fluids and antiemetics. These may be two areas of further investigation in the future. Patients were followed through delivery; there were no adverse maternal or fetal effects. In conclusion, total peripheral parenteral nutrition offers major advantages over conventional total parenteral nutrition to pregnant patients needing nutritional support. It is well tolerated, has minimal side effects, and is simple and cost-effective. Our study demonstrates that providing hypercaloric, hyperosmolar solutions via peripheral veins over short periods of time is feasible, especially in pregnant patients. Further studies using a control group need to be done to evaluate clinical effectiveness with respect to complications and subsequent pregnancy outcome. REFERENCES 1. Dudrick
SJ, Wilmore DN, Vars HM, et al: Long term total parenteral nutrition with growth development, and positive nitrogen
5. Stellato TA, Danziger LH, Burkons D. et al: Fetal salvage with maternal total parenteral nutrition: The pregnant mother as her own control. JPEN 12:412-413, 1988 6. O’Keefe SJD, Bean E, Symmonds K. et al: Clinical evaluation of a "3 in 1" intravenous nutrient solution. S Afr J Med 68:82-86, 1985 7. Wilson A, Goode AW, Kirk CJC, et al: Parenteral nutrition via peripheral veins: A feasibility study. J R Soc Med 80:430-433, 1987 8. Bodky A: Parenteral nutrition by peripheral vein central venous
9.
catheter or portal vein? World J Surg 10 :47-52, 1986 Gassanega A, Day AT, Sankary H: Efficacy of a 20% fat emulsion administered as a peripherally administered substrate. Surg Gynecol Obstet 160:387-392, 1985.
10.
11. 12.
Sayeed FA, Tripp MG, Sukumaran KB, et al: Stability of various total nutrient admixture formulations using Liposyn II and Aminosyn II. Am J Hosp Pharm 44:2280-2286, 1987 Flanagan S: Research highlights. Research Resources Reporter Feb:11-13, 1988 Gazitria R, Wilson K, Bistrain BR, et al: Factors determining peripheral vein tolerance to amino acid infusions. Arch Surg 114:897-900, 1979
13.
Fujiwara T, Kawarasaki H, Fonkalsrud E: Reduction of postinfusion venous endothelial injury with intralipid. Surg Gynecol
Obstet 158:57-65, 1984 14. Tanner WA, Delaney PV, Hennessey TP: The influence of heparin on intravenous infusions. Br J Surg 67:311-312, 1980 15. Maki DG: Preventing infection in intravenous therapy. Anesth Analg 56:141-153, 1977
Downloaded from pen.sagepub.com at NANYANG TECH UNIV LIBRARY on May 22, 2015
